PROBLEM with FLASH test problem

[Yên Bình]

Dear KROME users,

I tried to test KROME in FLASH 4.7.1. I try with the available primordial network. I copied all the file in Driver, physics and Simulation to FLASH. But I got some problem when I tried to compile the test problem Chemistry_Krome_Collapse.

/software/opt/mpich/4.0.1/intel/2020u4/bin/mpif90 -ggdb -c -O2   -DMAXBLOCKS=1000 -DNXB=8 -DNYB=8 -DNZB=1 -DN_DIM=2 KromeChemistry.F90
KromeChemistry.F90(105): error #6633: The type of the actual argument differs from the type of the dummy argument.   [XN]
      xn(krome_idx_e) = krome_get_electrons(xn(:))
----------------------------------------------------^
KromeChemistry.F90(108): error #6633: The type of the actual argument differs from the type of the dummy argument.   [XN]
              call krome(xn,tmp,dt)
-------------------------^
KromeChemistry.F90(108): error #6633: The type of the actual argument differs from the type of the dummy argument.   [TMP]
              call krome(xn,tmp,dt)
----------------------------^
KromeChemistry.F90(108): error #6633: The type of the actual argument differs from the type of the dummy argument.   [DT]
              call krome(xn,tmp,dt)
--------------------------------^
KromeChemistry.F90(111): error #6633: The type of the actual argument differs from the type of the dummy argument.   [XN]
      xn(krome_idx_e) = krome_get_electrons(xn(:))
----------------------------------------------------^
compilation aborted for KromeChemistry.F90 (code 1)
make: *** [Makefile:116: KromeChemistry.o] Error 1

I would be grateful to any ideas that help me solve this issue.

Thank you all.,

Binh

[tommaso grassi]

Hi Binh,

thanks for the email.

It seems that the compiler is assuming single precision for real declarations.

option -real-size 64, would help or either replace "real" declarations with "real*8".

Let me know if it works.

Thanks for using KROME.

cheers,

-tg